Mangrove Expansion Into Salt Marshes Alters Associated Faunal Communities

Home , Salt marsh

See discussions, stats, and author profiles for this publication at: https://www.researchgate.net/publication/313358137

Mangrove expansion into salt marshes alters associated faunal communities

Article · February 2017 DOI: 10.1016/j.ecss.2017.02.005

CITATIONS READS 0 7

4 authors, including:

Delbert 'Lee' Smee James A. Sanchez Texas A&M University - Corpus Christi Texas A&M University - Corpus Christi

46 PUBLICATIONS 567 CITATIONS 4 PUBLICATIONS 3 CITATIONS

SEE PROFILE SEE PROFILE

Meredith Diskin Texas A&M University - Corpus Christi

1 PUBLICATION 0 CITATIONS

SEE PROFILE

All content following this page was uploaded by Meredith Diskin on 27 February 2017.

The user has requested enhancement of the downloaded file. All in-text references underlined in blue are added to the original document and are linked to publications on ResearchGate, letting you access and read them immediately. Estuarine, Coastal and Shelf Science 187 (2017) 306e313

Contents lists available at ScienceDirect

Estuarine, Coastal and Shelf Science

journal homepage: www.elsevier.com/locate/ecss

Mangrove expansion into salt marshes alters associated faunal communities

* Delbert L. Smee a, , James A. Sanchez a, Meredith Diskin a, Carl Trettin b a Texas A&M University e Corpus Christi, Department of Life Sciences, 6300 Ocean Dr, Corpus Christi, TX 78412, United States b US Forest Service Southern Research Station, Cordesville, SC 29434, United States article info abstract

Article history: Climate change is altering the distribution of foundation species, with potential effects on organisms that Received 6 April 2016 inhabit these environments and changes to valuable ecosystem functions. In the Gulf of Mexico, black Received in revised form mangroves (Avicennia germinans) are expanding northward into salt marshes dominated by Spartina 26 January 2017 alterniflora (hereafter Spartina). Salt marshes are essential habitats for many organisms, including Accepted 3 February 2017 ecologically and economically important species such as blue crabs (Callinectes sapidus) and Penaeid Available online 4 February 2017 shrimp (e.g., Penaeus aztecus), which may be affected by vegetation changes. Black mangroves occupied higher tidal elevations than Spartina, and Spartina was present only at its lowest tidal elevations in sites Keywords: Spartina alterniflora when mangroves were established. We compared nekton and infaunal communities within monoculture Avicennia germinans stands of Spartina that were bordered by mangroves to nearby areas where mangroves had not yet Vegetation shift become established. Nekton and infaunal communities were significantly different in Spartina stands Climate change bordered by mangroves, even though salinity and temperature were not different. Overall abundance Shrimp and biomass of nekton and infauna was significantly higher in marshes without mangroves, although crabs and fish were more abundant in mangrove areas. Black mangrove expansion as well as other ongoing vegetation shifts will continue in a warming climate. Understanding how these changes affect associated species is necessary for management, mitigation, and conservation. © 2017 Elsevier Ltd. All rights reserved.

1. Introduction Along the eastern and Gulf of Mexico coasts of the US, salt marshes are abundant and provide numerous ecosystem services Foundation species are critically important for the structure and including shoreline protection, carbon and nutrient cycling, and function of communities and for ecosystem processes such as car- essential habitat for many species (Cuddington et al., 2011). At low bon cycling and energy flow (Ellison et al., 2005). Landscape level tidal elevations, salt marshes are dominated by Spartina alterniflora shifts in the distribution and abundance of foundation species can (hereafter Spartina), an essential foundation species, that creates fundamentally alter ecosystems (Armitage et al., 2015), and are critical habitat for many ecologically and economically important presently occurring with poleward vegetation species shifts caused species (e.g., blue crabs, Callinectes sapidus and red drum, Sciaenops by climate change (Micheli et al., 2008; Osland et al., 2013; Verges ocellatus), and serves as a significant detrital input that forms the et al., 2014) in both aquatic and terrestrial ecosystems (Gonzalez basis for coastal food webs (Rozas and Zimmerman, 2000; Pennings et al., 2010). For example, in North Carolina, eel grass (Zostera and Bertness, 2001; Simas et al., 2001; Stunz et al., 2002). Several marina) is being displaced by shoal grass (Halodule wrightii), other salt-tolerant species are present at higher tidal elevations reducing biodiversity in these areas (Micheli et al., 2008). A similar (e.g., Spartina patens, Batis maritima, Salicornia virginica) and also transition is underway in Texas, where shoal grass is being dis- provide habitat and shoreline protection. Primary production by placed by the tropical seagrasses Syringodium filiforme (manatee marsh plants is critical for associated fauna, and detrital inputs can grass) and Thalassia testudinum (turtle grass), significantly changing influence secondary production in adjacent systems (Peterson and seagrass-associated fauna (Ray et al., 2014). Howarth, 1987; Pennings and Bertness, 2001). In tropical climates, Spartina is outcompeted by mangrove trees that are well adapted to coastal environments (Reef et al., 2010; * Corresponding author. Simpson et al., 2013), and mangrove forests replace salt marshes E-mail address: [email protected] (D.L. Smee). http://dx.doi.org/10.1016/j.ecss.2017.02.005 0272-7714/© 2017 Elsevier Ltd. All rights reserved. D.L. Smee et al. / Estuarine, Coastal and Shelf Science 187 (2017) 306e313 307 as the primary coastal wetland. Recent studies have documented a but, not in the north. Our study marshes were intertidal, experi- northward migration of black mangroves (Avicennia germinans), encing similar tidal fluctuations (~0.5 m) that are influenced by attributed to a reduction in severe freezing events over the past 3 diurnal and semi-diurnal tides but primarily by prevailing south- decades (Cavanaugh et al., 2014). Spartina also faciliates mangrove east winds. Faunal samples were collected during periods of high expansion by trapping seedlings in suitable growth areas (Peterson monthly tides in June of 2013 (10e11 and 24e25) when the highest and Bell, 2012) and by creating a warmer layer that surrounds marsh and mangrove elevations were submerged. Water temper- seedlings, protecting them from cold temperatures (Guo et al., ature was 29 C ± 0.9 C and 30 C ± 1.2, and salinity was 2013). Historically in the Western Gulf of Mexico, particularly in 32 ± 2.3 ppt and 34 ± 1.8 ppt in sites with and without mangroves Texas, black mangroves were present, but populations periodically respectively. expanded, displacing other marsh plants, and then contracted allowing those plants to reemerge in response to variations in 2.1. Vegetation survey climate, but black mangroves did not become established in subtropical climates that experienced winter freezes (Sherrod and We measured relative tidal elevations of Spartina and black McMillan, 1981; Sherrod and McMillan, 1985; McMillan and mangroves. Spartina will not grow in subtidal areas. Therefore, we Sherrod, 1986; Cavanaugh et al., 2014). However, the frequency of used the boundary between Spartina and benthic habitats (mud severe cold weather events has declined, and black mangroves have flats or seagrass beds in this area) as our lowest elevation point (i.e. migrated northward and expanded their distribution in the Gulf of point zero). Then, using a laser level we calculated the tidal ele- Mexico. In the Mission-Aransas Estuary near Aransas Pass, Texas, vations in which Spartina and mangroves were found relative to USA, only 65 acres of black mangrove forest were reported in the point zero along a transect from the lowest tidal elevation land- 1980s. From 1989 to 2005, black mangroves expanded and are ward. In marshes where black mangroves were established and estimated to cover between 15,000 and 21,500 acres (Montagna abundant, mangroves were present at higher elevations above et al., 2007). Expanding mangrove populations in Texas have dis- mean lower low water while Spartina was present only in a narrow placed Spartina and other plants (e.g., S. virginica, B. maritima)in band ~1e4 m wide at the lowest tidal elevations (closest to mean coastal marshes (Everitt et al., 2010; Armitage et al., 2015). lower low water). In contrast, Spartina dominated at both low and Salt marshes are among the most productive ecosystems on high tidal elevations in areas without abundant mangroves (Fig. 1, earth, and their detrital input is an important resource for many see Results). Spartina elevations are reported from sites without species in the marsh and in adjacent communities (Pennings and mangroves because Spartina elevations are compressed in Bertness, 2001; Simas et al., 2001). Like marshes, mangrove forests are productive and support a diversity of fauna that are ecologically and economically important (Vaslet et al., 2012). However, there are considerable differences in the composition and allocation of biomass among marshes and mangroves. Standing biomass is greater in mangroves than marshes, while marshes tend to exhibit higher net primary productivity (Alongi, 1998; Alongi et al., 2004). Organic matter turnover is lower in mangrove forests than marshes because ~60% of the mangrove biomass is woody (Alongi, 1998; Castaneda-Moya~ et al., 2013). Soil conditions and biogeochemical processes also differ between these habitats (Patterson and Mendelssohn, 1991; Perry and Mendelssohna, 2009; Comeaux et al., 2012), however those differences may not be evident soon after a change in plant community composition (Henry and Twilley, 2013). Despite the noted differences in detrital production, biomass, and soil processes, few studies have assessed the impacts of mangrove encroachment into marshes on associated fauna, although two recent studies suggest that the effects on associated nekton and benthic fauna are likely to be significant (Caudill, 2005; Lunt et al., 2013). The northward migration of mangroves in the Gulf Coast has been occurring for the past 25þ years and if populations continue to expand their range, replacement of Spartina as the primary coastal wetland species may occur (Montagna et al., 2007; Osland et al., 2013), but the effects on associated fauna remain obscure. Here, we measured the tidal elevation distributions of Spartina and black mangroves in the Western Gulf of Mexico. Then, we collected nekton and infaunal organisms at two tidal elevations in marshes dominated by black mangroves and in marshes where black mangroves were rare. These marshes were separated by <10 km and had similar abiotic conditions, allowing us to assess the effects on mangrove encroachment on associated fauna. Fig. 1. Marshes with black mangroves (top) and without black mangroves (bottom). 2. Methods Black mangroves were present only at higher tidal elevations in a near monoculture once established. Spartina dominated at low and high elevations when mangroves were not present. Nekton and infaunal samples were collected from both tidal heights This study was performed in the Mission-Aransas National in both types of marshes, and the tidal elevation of Spartina and black mangroves was Estuarine Research Reserve (MANERR) near Rockport, Texas, USA. measured. In this figure, higher tidal elevations are on the left and the arrows repre- Mangroves are established in the southern part of the MANERR, sent paired sampling locations in each marsh type. 308 D.L. Smee et al. / Estuarine, Coastal and Shelf Science 187 (2017) 306e313 mangrove sites. using a 500 mm sieve and transferred to jars containing 45% isopropyl alcohol. Infaunal organisms were sorted from detritus and 2.2. Faunal collections plant material using a dissecting microscope. Polychaetes were identified to family and all other organisms were identified to the We classified marshes into two types: those with abundant lowest taxonomic level possible for use in multivariate analyses. black mangroves and those where black mangroves were rare. Black mangroves were found in higher tidal elevations, and to ac- 2.3. Data analysis count for both mangrove presence and tidal elevation in our sampling, our collections were made in a 2 2 factorial design, with Nekton and infaunal samples were examined separately using mangrove presence and tidal elevation as factors. Samples at each univariate analyses in JMP Pro 12 and multivariate analyses in e elevation were paired and were ~ 2 3 m apart. We collected PRIMER ™. For nekton, the total abundance of fish, crabs, and grass community samples at 2 tidal elevations (~0.1 m and ~0.4 m above shrimp (Palaemonetes spp.), as well as the biomass of each group mean lower low water) in 15 sites with and 14 sites without were compared using a two-way ANOVA with mangrove presence mangroves (58 samples total). Our intent was collect 15 paired (present/absent) and tidal elevation (low/high) as fixed factors. For samples in each marsh type, but our sampler malfunctioned at the infauna, total abundance of the two most abundant groups: poly- last site, leaving us with 14 paired samples in marshes without chaetes and crustaceans were similarly analyzed using a two-way mangroves. We used the lowest tidal elevation that Spartina was ANOVA with mangrove presence (present/absent) and tidal eleva- present as an estimate of mean lower low water because Spartina tion (low/high) as fixed factors. Data were log transformed to meet will not grow subtidal. Sampling sites were separated by more than assumptions of equal variances. For both nekton and infauna, we 150 m to cover a large spatial area. In marshes with established created metric, 2D MDS plots from mean values of communities in mangroves, samples in lower elevations were collected within each sample. We also performed multivariate analyses on nekton Spartina and higher elevations samples were always collected and infaunal community assemblages within each sample using within mangrove pneumatophores. In marshes without man- Analysis of Similarity (ANOSIM) in PRIMER ™ with mangrove groves, samples at both low and high tidal elevations were presence as the factor. SIMPER analysis was also performed to collected within Spartina (Fig. 1). Water depth was recorded at each determine which organisms contributed to dissimilarities between fi fl sampling location, but, did not signi cantly in uence community mangrove and Spartina habitats. assemblages and is not reported. All faunal samples were collected during highest seasonal tides to ensure the highest tidal elevations were submerged. 3. Results

2.2.1. Nekton communities 3.1. Vegetation survey Nekton were collected using a suction sampler, which has been successfully used in previous coastal wetland studies (Heck et al., Spartina was present at lower tidal elevations than black man- 1995). A 0.25 m2 cylindrical plastic barrel (open at both ends) groves (median 0.35 m vs. 0.44 m), although mangroves were found was placed over a selected salt marsh or black mangrove habitat over a larger range of tidal elevations (Fig. 2). In marshes sites and was firmly pressed into the substrate to prevent organisms without mangroves, Spartina was found in monocultures at both from escaping. The barrel encapsulated either Spartina shoots or low and high elevations. In contrast, when black mangroves were black mangrove pneumatophores, allowing us to collect organisms abundant, Spartina was relegated to tidal elevations below ~0.2 m, directly within that particular vegetation type. After the barrel was and found in a narrow band bordering the dwarf mangrove forest placed, water was suctioned for three minutes to ensure all or- (Fig. 1). Areas with equal mixtures of mangroves and Spartina at the fi fi ganisms trapped within the barrel were collected. Preliminary same tidal elevations were dif cult to nd and were not sampled. sampling indicated that 3 min was sufficient time for collecting all organisms. Samples were placed into 250 mm mesh bags, labeled, 3.2. Nekton communities and placed on ice. Upon returning from the field, samples were placed into buckets containing 10% formalin for one week, and then Grass shrimp were significantly more abundant and had more transferred to jars containing 45% isopropyl alcohol. All organisms biomass in marsh sites without mangroves (p < 0.01, Fig. 3, Tables 1 were identified to the lowest taxonomic level possible for use in and 2), but fish were significantly more abundant and more fish multivariate analyses. Total abundance of fish, shrimp, and crabs biomass was found in marsh sites that had abundant mangroves was calculated. Dry weight biomass was determined for all fish, shrimp, and crabs collected by drying samples at 90 C for 72 h, and weighing them on an analytical balance to the nearest 0.001 g. All 1.25 individual fish were weighed together for each sample, as were 1 shrimp and crabs, to create a single measure of biomass for that group. 0.75

0.5

2.2.2. Infaunal communities meters Infaunal organisms were collected using PVC cores. Two sedi- 0.25 ment cores (10 cm diameter x 10 cm depth) were taken at both low and high tidal elevations (4 total cores) in 5 sites with and 5 sites 0 without mangroves. The number of organisms in cores taken at the SparƟna Mangrove same tidal elevation were combined for analysis. Benthic cores were placed into 250 mm mesh bags, gently rinsed with seawater in Fig. 2. Box and whiskers plot for tidal elevations of Spartina alterniflora and black mangroves. Boxes show median with 25% and 75% quartiles, and whiskers are range of the field to remove excess sediment, labeled, and placed on ice. heights that each type of vegetation was found. Spartina heights are reported only for fi Upon returning from the eld, samples were placed into buckets sites without mangroves because mangroves displace Spartina at its highest tidal containing 10% formalin for a period of one week, and then washed elevations. D.L. Smee et al. / Estuarine, Coastal and Shelf Science 187 (2017) 306e313 309

Grass Shrimp Abundance Table 2 ANOVA results for biomass of shrimp, crabs, and ﬁsh. Signiﬁcant effects (p < 0.05) 50 are indicated in bold print. 40 Grass shrimp biomass (F ¼ 5.233,52,P< 0.001) 30 Source Sum of squares df F P

20 Mangrove Presence 1.03 1 20.9 <0.001 10 Tidal Elevation >0.001 1 0.02 0.89 Mangrove Presence*Tidal Elevation >0.001 1 0.001 0.99 0 ¼ ¼ Mangrove SparƟna Crab biomass (F 1.173,43, P 0.33) Source Sum of squares df F P Low High Mangrove Presence 0.32 1 0.89 0.35 Crab Abundance Tidal Elevation 0.87 1 2.4 0.12 Mangrove Presence*Tidal Elevation 0.24 1 0.66 0.41 15 Fish biomass (F ¼ 3.173,33, P ¼ 0.04) Source Sum of squares df F P 10 Mangrove Presence 1.99 1 8.77 0.006 Tidal Elevation 0.003 1 0.01 0.90 5 Mangrove Presence*Tidal Elevation 0.15 1 0.66 0.42

Mean Abundance 0 fi Mangrove SparƟna not a signi cant factor for any of the comparisons of grass shrimp, crab, and fish abundance or biomass for these organisms among Low High sites (Fig. 3, Tables 1 and 2). The interaction between tidal elevation and mangrove presence was not significant for crab and fish Fish Abundance abundance nor for biomass of any of these groups, but the inter- 5 action term was significant (P ¼ 0.55) for grass shrimp abundance 4 (Tables 1 and 2) (see Figs. 3 and 4). A metric MDS plot revealed that the greatest differences among 3 samples were from marshes with and without mangroves and not 2 from low and high tidal elevations, which was consistent with 1 univariate analysis on abundances and biomass of grass shrimp, fi fi 0 crabs, and sh. Because tidal elevation did not have a signi cant Mangrove SparƟna effect and had a smaller effect on the community than mangrove presence, ANOSIM was performed using mangrove presence as a Low High factor influencing community differences. ANOSIM revealed sig- fi Fig. 3. Mean þ SE abundances of shrimp, crabs, and fish in samples from high and low ni cant effects of mangrove presence on overall community as- tidal elevations from marshes with and without mangroves. Shaded bars are high tidal semblages (R ¼ 0.101, p ¼ 0.001). However, R values approaching 0, elevations, and x axis labels are marshes with abundant mangroves vs. those with only such as the 0.101 value found here, suggests high variability within fi fi a ¼ fi Spartina. Signi cant differences were found for shrimp and sh at 0.05. Signi cant the data set. Since nekton are highly mobile and capable of moving differences between tidal elevations were not found. among areas, this variability is not unexpected. The greatest dissimilarity between communities was attributed to grass shrimp, Table 1 brown shrimp (Penaeus aztecus), and blue crabs that were more ANOVA results for abundances of shrimp, crabs, and fish. Significant effects (p < abundant in areas without mangroves and mud crabs (Panopeidae, 0.05) are indicated in bold print. Xanthidae) that were more abundant in mangrove areas (Table 3).

Grass shrimp abundance (F ¼ 4.033,52,P¼ 0.01)

Source Sum of squares df F P 3.3. Infaunal communities Mangrove Presence 2.47 1 8.34 0.006 Tidal Elevation 0.002 1 0.01 0.93 Polychaetes were significantly more abundant in marsh sites Mangrove Presence*Tidal Elevation 1.15 1 3.88 0.055 without mangroves (F1, 1 ¼ 4.43, p ¼ 0.06), while tidal elevation did Crab abundance (F ¼ 4.03 , P ¼ 0.32) 3,43 not significantly affect polychaete abundance (F1, 1 ¼ 0.67, p ¼ 0.43 Source Sum of squares df F P Fig. 5), and the interaction between mangrove presence and tidal fi ¼ ¼ Mangrove Presence 0.99 1 0.77 0.38 height was not signi cant (F1, 1 0.55, p 0.47). Over 95% of the Tidal Elevation 2.78 1 2.15 0.15 crustaceans collected were from the Tanaididae family. Like poly- Mangrove Presence*Tidal Elevation 1.68 1 1.30 0.26 chaetes, crustaceans, were significantly more abundant in marsh ¼ ¼ Fish abundance (F ¼ 4.033,33, P ¼ 0.04) sites without mangroves (F1, 1 7.24, p 0.03). Tidal elevation did not significantly affect crustacean abundance (F ¼ 0.14, p ¼ 0.71 Source Sum of squares df F P 1, 1 Fig. 5), and the interaction between mangrove presence and tidal Mangrove Presence 4.79 1 9.25 0.005 height was not significant (F1, 1 ¼ 0.01, p ¼ 0.97). Similar to nekton Tidal Elevation 0.08 1 0.15 0.70 fi Mangrove Presence*Tidal Elevation 0.06 1 0.13 0.72 ndings and consistent with ANOVA results, a metric MDS plot revealed that the greatest differences among infaunal communities were from samples taken from marshes with and without man- < (p 0.01, Fig. 3, Tables 1 and 2). Crab abundance and biomass did groves and not between tidal elevations (Fig. 6). ANOSIM revealed fi not differ signi cantly between marshes with and without man- significant effects of mangrove presence on infaunal community ¼ ¼ groves (p 0.38, p 0.35, Fig. 3, Tables 1 and 2). Tidal elevation was assemblages (R ¼ 0.62, p ¼ 0.001). The global R value for mangrove 310 D.L. Smee et al. / Estuarine, Coastal and Shelf Science 187 (2017) 306e313

Metric MDS Polychaetes Resemblance: S17 Bray-Curtis similarity 25 2D Stress: 0.01 20 20 Yes/LowMG + Low 15 Yes/HighMG + High No/LowMG - Low 10 No/HighMG - High 5 0 0 MDS2 Crustaceans

Mean Abundance 40

-20 20

-20 0 20 10 MDS1 0 Transform: Square root Mangrove SparƟna Resemblance: S17 Bray-Curtis similarity Low High 2D Stress: 0.25 Fig. 5. Mean þ SE abundances of polychaetes and crustaceans from infaunal samples 50 collected in high and low tidal elevations from marshes with and without mangroves. Shaded bars are high tidal elevations, and x axis labels are marshes with abundant mangroves vs. those with only Spartina. Significantly more polychaetes and crustaceans were found in marshes without mangroves at a ¼ 0.05. Significant differences 0 between tidal elevations were not found. S2 D M species in coastal wetlands (Gonzalez et al., 2010; Barbier et al., -50 2011; Comeaux et al., 2012), often with profound impacts on ecosystem processes and associated fauna (Micheli et al., 2008; Armitage et al., 2015). Invasion by the common reed (Phragmites australis) decreased faunal abundance and diversity in US Atlantic -100 salt marshes (Osgood et al., 2003; Kimball et al., 2010) as did -150 -100 -50 0 50 100 changes in seagrass species in North Carolina (Micheli et al., 2008) MDS1 and Texas (Ray et al., 2014). Expansion of mangroves into salt marshes has occurred in North and South America and Australia, Fig. 4. Metric MDS plot showing (A) mean and (B) all distances between nekton decreasing marsh habitat (Saintilan et al., 2014). In the Gulf of samples from marshes with and without mangroves at low and high tidal elevations. Black triangles indicate mangroves were present (MGþ), and white triangles indicate Mexico, black mangroves are displacing Spartina and other marsh they were absent (MG-). Triangles point up for high elevation and down for low plants (Armitage et al., 2015), altering wetland elevations and elevation. ecosystem processes such as nutrient cycling (Comeaux et al., 2012). Our findings indicate that changes from Spartina dominated to effects on infaunal communities was much higher than found for mangrove dominated coastal wetlands will have significant im- nekton communities, probably because infaunal communities are pacts on associated fauna in the Gulf of Mexico. Infaunal commu- far less mobile than nekton. The greatest dissimilarity between nities, blue crabs, and both grass and brown shrimp were less communities was attributed to Tanaidae, Nereididae, Amphipods, abundant in marshes with abundant mangroves, even when and Capitellidae (Table 4). collected from low tidal elevations where Spartina was present. In contrast, mud crabs and fish were more abundant in marshes 4. Discussion dominated by mangroves. Similarly, Lunt et al. (2013) found infauna to be lower in mangrove areas. In Louisiana, mangrove effects on Climate change and invasion of exotic species, which in some nekton differed from those reported here (Caudill, 2005), with fish cases are facilitated by a warming climate, are altering foundation

Table 3 SIMPER analysis for nekton. Abundances shown were square root transformed for analysis.

Species Mangrove Spartina Mangrove vs. Spartina

Mean abundance Mean abundance Dissimilarity % Contributed

Grass shrimp (Palaemonetes) 2.71 4.74 19.3 29.8 Mud crabs (Xanthidae/Panopeidae) 1.59 1.05 9.9 15.3 Brown shrimp (Penaeus aztecus) 0.58 0.91 5.9 9.1 Blue crabs (Callinectes sapidus) 0.39 0.48 3.6 5.6 D.L. Smee et al. / Estuarine, Coastal and Shelf Science 187 (2017) 306e313 311

Metric MDS areas that were dominated by other plant species (i.e. Spartina). We Resemblance: S17 Bray-Curtis similarity do not yet know how mangroves are inﬂuencing wetland species but hypothesize that mangroves alter food webs through changes 2D Stress: 0.06 XXXXXXXXXXX Yes-LowXXXXXXXXX in the outcomes of predatory interactions (Hammerschlag et al., XXXXXXXXXXXXXXXXXXXXNo-Low 2010) and/or by affecting abiotic conditions and resource avail- 20 ability to make habitats less suitable for some species (Alongi et al., Yes-LowMG + Low 2004; Castaneda-Moya~ et al., 2013; Comeaux et al., 2012). Yes-HighMG + High Black mangrove edge habitats are used by predators as hunting corridors (Hammerschlag et al., 2010), and mangrove encroach- No-LowMG - Low 0 ment into Spartina marshes may alter food webs by changing the

MDS2 MG - High No-High foraging environment for predators and prey. In sites with mangroves, fish and crabs were more abundant while infaunal organ- XXXXXXXXXXXNo-High isms and grass shrimp, common prey items for fish and crabs, Yes-HighXXXXXXXXXXX XXXXXXXXX -20 XXXXXXXXX (Kneib and Wagner, 1994), were less abundant. Mangroves accu- mulate sediment, raising the tidal elevation and reducing emer- gence time for the areas where they are found (Comeaux et al., -40 -20 0 20 40 2012). Thus, common wetland aquatic species would spend less MDS1 time in the highest tidal elevations and more time in the fringing Transform: Square root Spartina marshes. By altering the landscape, mangroves may place Resemblance: S17 Bray-Curtis similarity predators and prey in close proximity and concentrate them by 50 2D Stress: 0.21 reducing suitable habitat area, thereby altering encounter rates, predation levels, and food web structure. Additionally, resource availability via detrital inputs is vastly different from woody mangroves and Spartina, and likely influence 0 358 food webs through bottom-up forces (Alongi, 1998; Alongi et al., 2004; Castaneda-Moya~ et al., 2013). Geochemical processes, sedi- 359 ment accretion, and carbon sequestration and turnover differ

MDS2 significantly among black mangrove and Spartina habitats (Alongi 360 et al., 2004; Castaneda-Moya~ et al., 2013; Comeaux et al., 2012), -50 and the distribution of organisms, particularly infaunal species, are often influences by changes in these factors (Baguley et al., 2006). Mangroves may simultaneously influence both top-down and bottom-up factors, thereby influencing the diversity and abun- -100 -50 0 50 100 dance of associated fauna. Clearly more research is needed to MDS1 elucidate the mechanisms by which mangroves affect associated fauna. Fig. 6. Metric MDS plot showing (A) mean and (B) all distances between infaunal Tidal elevation and inundation often have significant effects on samples from marshes with and without mangroves at low and high tidal elevations. fl Black triangles indicate mangroves were present (MGþ), and white triangles indicate coastal communities, in uencing nekton abundance in coastal they were absent (MG-). Triangles point up for high elevation and down for low wetlands (Rozas and Minello, 1998; Rozas and Zimmerman, 2000). elevation. Brown shrimp and grass shrimp can be more abundant at lower tidal elevations (Minello et al., 1994), but in our study, tidal elevation did not significantly affect community assemblages or abun- more abundant in marsh sites and shrimp more abundant in dances of individual organisms. Consistent with our findings, mangrove areas. Caudill's differing findings may be related to communities in Chesapeake Bay were not significantly different in methodology as they used lift nets to capture organisms on out- salt marshes between low and high tidal elevations (Posey et al., going tides and their community composition differed consider- 2003). We sampled at the lowest tidal elevations possible that ably. We found few gulf killifish Fundulus grandis and no white would allow us to collect organisms in both Spartina and mangrove shrimp Litopenaeus setiferus, but, both were abundant in Caudill's areas, and perhaps the differences in tidal elevations were not large (2005) study. Further, grass shrimp were abundant in our study enough to observe differences seen in other studies (Rozas and but and rarely collected by Caudill (2005), perhaps due to variation Zimmerman, 2000). Regardless, the presence of mangroves influ- in sampling gear. Regardless of whether differences in these studies enced community composition at the lowest elevations that were result from different sampling methods or different communities in dominated by Spartina, independent of tidal height sampled, which the northern vs. southern Gulf of Mexico, both studies suggest provides strong evidence that mangroves are influencing wetland mangroves affect faunal composition. Our findings indicate fauna. mangrove presence influenced associated marsh fauna in adjacent

Table 4 SIMPER analysis for infauna. Abundances shown were square root transformed for analysis.

Species Mangrove Spartina Mangrove vs. Spartina

Mean abundance Mean abundance Dissimilarity % Contributed

Tanaidae 0.14 2.69 23.4 27.9 Nereididae 1.23 1.28 14.5 17.3 Amphipod 0.34 1.51 13.2 15.8 Capitellidae 0 1.04 10.2 12.2 312 D.L. Smee et al. / Estuarine, Coastal and Shelf Science 187 (2017) 306e313

Vegetation shifts are occurring across the globe in response to rising sea levels. Estuar. Coast. Shelf Sci. 96, 81e95. http://dx.doi.org/10.1016/ anthropogenic effects including climate change (Gonzalez et al., j.ecss.2011.10.003. Cuddington, K., Byers, J., Wilson, W., Hastings, A., 2011. Ecosystem Engineers: Plants 2010), which can alter fauna that utilize these habitats (Micheli to Protists. Academic Press. et al., 2008; Ray et al., 2014). Both nekton and infaunal commu- Ellison, A.M., Bank, M.S., Clinton, B.D., Colburn, E.A., Ford, C.R., Foster, D.R., nities were significantly different in locations where mangroves Kloeppel, B.D., Knoepp, J.D., Lovett, G.M., Mohan, J., Orwig, D.A., Rodenhouse, N.L., Sobczak, W.V., Kristina, A., Stone, J.K., Swan, C.M., became established, suggesting that mangrove expansion into Thompson, J., Holle, B.Von, Jackson, R., Ellisonl, A.M., Bank, M.S., Clinton, B.D., marsh areas previously dominated by Spartina will have significant Colburnm, E.A., Elliott, K., Ford, C.R., Foster, D.R., Kloeppel, B.D., Knoepp, J.D., effects on estuarine food webs and biodiversity. Our results provide Lovett, G.M., Mohan, J., Orwig, D.A., Rodenhouse, N.L., Sobczak, W.V., Stinson, K.A., Stone, J.K., Swan, C.M., Thompson, J., Holle, B.Von, Webster, J.R., a baseline for future studies examining effects of mangrove 2005. ecosystems. Front. Ecol. Environ. 3, 479e486. encroachment on fauna that inhabit coastal wetlands. The mech- Everitt, J., Yang, C., Summy, K., 2010. Use of archive aerial photography for 170 black anisms by which mangroves are affecting fauna and the broader mangrove populations. J. Coast. Res. 26, 649e653. Gonzalez, P., Neilson, R.P., Lenihan, J.M., Drapek, R.J., 2010. Global patterns in the ecosystem consequences of faunal changes requires further inves- vulnerability of ecosystems to vegetation shifts due to climate change. Glob. tigation, but understanding how these changes affect alter species Ecol. Biogeogr. 19, 755e768. http://dx.doi.org/10.1111/j.1466- diversity and abundance are necessary for management, mitiga- 8238.2010.00558.x. tion, and conservation of coastal wetlands and associated species. Guo, H.Y., Zhang, Y.H., Lan, Z.J., Pennings, S.C., 2013. Biotic interactions mediate the expansion of black mangrove (Avicennia germinans) into salt marshes under climate change. Glob. Chang. Biol. 19, 2765e2774. http://dx.doi.org/10.1111/ Author contributions gcb.12221. Hammerschlag, N., Morgan, A.B., Serafy, J.E., 2010. Relative predation risk for fishes along a subtropical mangrove-seagrass ecotone. Mar. Ecol. Prog. Ser. 401, DLS, JAS, and CT conceived the idea and designed the experi- 259e267. http://dx.doi.org/10.3354/meps08449. ment. DLS and CT acquired funding. DLS, JAS, and MD collected the Heck, K.L., Able, K.W., Roman, C.T., Fahay, M.P., 1995. Composition, abundance, biomass, and production of macrofauna in a New england estuary: comparisons samples and generated the data. DLS, JAS, MD, and CT analyzed the among eelgrass meadows and other nursery habitats. Estuaries 18, 379. http:// data, and then wrote and edited the manuscript. dx.doi.org/10.2307/1352320. Henry, K.M., Twilley, R.R., 2013. Soil development in a coastal Louisiana wetland during a climate-induced vegetation shift from salt marsh to mangrove. J. Coast. Acknowledgements Res. 292, 1273e1283. http://dx.doi.org/10.2112/JCOASTRES-D-12-00184.1. Kimball, M.E., Able, K.W., Grothues, T.M., 2010. Evaluation of long-term response of Funding was provided by the US Forest Service Southern intertidal creek nekton to phragmites australis (common reed) removal in oligohaline Delaware Bay salt marshes. Restor. Ecol. 18, 772e779. http:// Research Station agreements 12-DG-11330101-096 and 13-CA- dx.doi.org/10.1111/j.1526-100X.2009.00543.x. 11330140-116 to D.L. Smee. The NSF-MSP ETEAMS grant #1321319 Kneib, R.T., Wagner, S.L., 1994. Nekton use of vegetated marsh habitats at different provided funding for boat time and their interns assisted in the stages of tidal inundation. Mar. Ecol. Prog. Ser. 106, 227e238. http://dx.doi.org/ fi 10.3354/meps106227. eld. Members of the Marine Ecology Lab including J. Lunt and A. Lunt, J., McGlaunn, K., Robinson, E., 2013. Effects of black mangrove (Avicennia Scherer, and Julie Arnold and Christina Stringer from USFS provided germinans) expansion on salt marsh (Sparrtina alterniflora) benthic commu- important assistance in the field for both collections and mea- nities of the south Texas coast. Gulf Caribb. Res. 1, 125e129. McMillan, C., Sherrod, C., 1986. The chilling tolerance on black mangrove, Avicennia surements of vegetation tidal elevations. Students in TAMU-CC's germinans, from the Gulf of Mexico coast of Texas, Louisiana and Florida. REU program also provided field help. S. Bock was instrumental for Contrib. Mar. Sci. 29, 9e16. writing and data analysis. Micheli, F., Bishop, M.J., Peterson, C.H., Rivera, J., 2008. Alteration of seagrass species composition and function over two decades. Ecol. Monogr. 78, 225e244. http:// dx.doi.org/10.1890/06-1605.1. Appendix A. Supplementary data Minello, T., Zimmerman, R., Medina, R., 1994. The importance of edge for natant macrofauna in a created salt marsh. Wetlands 14, 184e198. Montagna, P.A., Gibeaut, J.C., Tunnell Jr., J.W., 2007. South Texas climate 2100: Supplementary data related to this article can be found at http:// coastal impacts. In: Norwine, J., John, K. (Eds.), The Changing Climate Of South dx.doi.org/10.1016/j.ecss.2017.02.005. Texas 1900-2100: Problems And Prospects, Impacts And Implications. Texas A&M University-Kingsville, pp. 57e77. Osgood, D.T., Yozzo, D.J., Chambers, R.M., Jacobson, D., Hoffman, T., Wnek, J., 2003. References Tidal hydrology and habitat utilization by resident nekton in Pragmites and Non-Phragmites Marshes. Estuaries 26, 522e533. http://dx.doi.org/10.1007/ Alongi, D.M., 1998. Coastal Ecosystem Processes. BF02823728. Alongi, D.M., Sasekumar, A., Chong, V.C., Pfitzner, J., Trott, L.A., Tirendi, F., Dixon, P., Osland, M.J., Enwright, N., Day, R.H., Doyle, T.W., 2013. Winter climate change and Brunskill, G.J., 2004. Sediment accumulation and organic material flux in a coastal wetland foundation species: salt marshes vs. mangrove forests in the managed mangrove ecosystem: estimates of land-ocean-atmosphere exchange southeastern United States. Glob. Chang. Biol. 19, 1482e1494. http://dx.doi.org/ in peninsular Malaysia. Mar. Geol. 208, 383e402. http://dx.doi.org/10.1016/ 10.1111/gcb.12126. j.margeo.2004.04.016. Patterson, C.S., Mendelssohn, I.A., 1991. A comparison of physicochemical variables Armitage, A., Highfield, W., Brody, S., Louchouarn, P., 2015. The contribution of across plant zones in a mangal/salt marsh community in Louisiana. Wetlands mangrove expansion to salt marsh loss on the Texas Gulf Coast. PLoS One 10, 11, 139e161. http://dx.doi.org/10.1007/BF03160845. e0125404. Pennings, S.C., Bertness, M.D., 2001. Salt marsh communities. In: Bertness, M.D., Baguley, J.G., Montagna, P.a., Hyde, L.J., Kalke, R.D., Rowe, G.T., 2006. Metazoan Gaines, S.D., Hay, M.E. (Eds.), Marine Community Ecology, pp. 289e316. meiofauna abundance in relation to environmental variables in the northern Perry, C.L., Mendelssohn, I.a., 2009. Ecosystem effects of expanding populations of Gulf of Mexico deep sea. Deep. Res. Part I Oceanogr. Res. Pap. 53, 1344e1362. Avicennia germinans in a Louisiana salt marsh. Wetlands 29, 396e406. http:// http://dx.doi.org/10.1016/j.dsr.2006.05.012. dx.doi.org/10.1672/08-100.1. Barbier, E., Hacker, S.D., Kennedy, C., Koch, E.W., Stier, A.C., Silliman, B.R., 2011. The Peterson, B.J., Howarth, R.W., 1987. Sulfur, carbon, and nitrogen isotopes used to value of estuarine and coastal ecosystem services. Ecol. Monogr. 81, 169e193. trace organic matter flow in the salt-marsh estuaries of Sapelo Island, Georgia. Castaneda-Moya,~ E., Twilley, R.R., Rivera-Monroy, V.H., 2013. Allocation of biomass Limnol. Oceanogr. 32, 1195e1213. http://dx.doi.org/10.4319/lo.1987.32.6.1195. and net primary productivity of mangrove forests along environmental gradi- Peterson, J.M., Bell, S.S., 2012. Tidal events and salt-marsh structure influence black ents in the Florida Coastal Everglades, USA. For. Ecol. Manage 307, 226e241. mangrove ( Avicennia germinans ) recruitment across an ecotone. Ecology 93, http://dx.doi.org/10.1016/j.foreco.2013.07.011. 1648e1658. http://dx.doi.org/10.1890/11-1430.1. Caudill, M.C., 2005. Nekton Utilization of Black Mangrove (AVicennia Germinans) Posey, M.H., Alphin, T.D., Meyer, D.L., Johnson, J.M., 2003. Benthic communities of and Smooth Cordgrass (Spartina alterniflora) Sites in Southwest Louisiana. MS common reed Phragmites australis and marsh cordgrass Spartina alterniflora Thesis. Louisana State University. marshes in Chesapeake Bay. Mar. Ecol. Prog. Ser. 261, 51e61. http://dx.doi.org/ Cavanaugh, K.C., Kellner, J.R., Forde, A.J., Gruner, D.S., Parker, J.D., Rodriguez, W., 10.3354/meps261051. Feller, I.C., 2014. Poleward expansion of mangroves is a threshold response to Ray, B.R., Johnson, M.W., Cammarata, K., Smee, D.L., 2014. Changes in seagrass decreased frequency of extreme cold events. Proc. Natl. Acad. Sci. 111, 723e727. species composition in northwestern Gulf of Mexico estuaries: effects on http://dx.doi.org/10.1073/pnas.1315800111. associated seagrass fauna. PLoS One 9, e107751. http://dx.doi.org/10.1371/ Comeaux, R.S., Allison, M.A., Bianchi, T.S., 2012. Mangrove expansion in the Gulf of journal.pone.0107751. Mexico with climate change: implications for wetland health and resistance to Reef, R., Feller, I.C., Lovelock, C.E., 2010. Nutrition of mangroves. Tree Physiol. 30, D.L. Smee et al. / Estuarine, Coastal and Shelf Science 187 (2017) 306e313 313

1148e1160. http://dx.doi.org/10.1093/treephys/tpq048. enrichment on Avicennia germinans in the salt marsh-mangrove ecotone. Rozas, L., Zimmerman, R.J., 2000. Small-scale patterns of nekton use among marsh Aquat. Bot. 104, 55e59. http://dx.doi.org/10.1016/j.aquabot.2012.09.006. and adjacent shallow nonvegetated areas of the Galveston Bay Estuary, Texas Stunz, G., Minello, T., Levin, P., 2002. A comparison of early juvenile red drum (USA). Mar. Ecol. Prog. Ser. 193, 217e239. densities among various habitat types in Galveston Bay, Texas. Estuaries 25, Rozas, L.P., Minello, T.J., 1998. Nekton use of salt marsh, seagrass, and nonvegetated 76e85. habitats in a South Texas (USA) estuary. Bull. Mar. Sci. 63, 481e501. Vaslet, a., Phillips, D.L., France, C., Feller, I.C., Baldwin, C.C., 2012. The relative Saintilan, N., Wilson, N.C., Rogers, K., Rajkaran, A., Krauss, K.W., 2014. Mangrove importance of mangroves and seagrass beds as feeding areas for resident and expansion and salt marsh decline at mangrove poleward limits. Glob. Chang. transient ﬁshes among different mangrove habitats in Florida and Belize: evi- Biol. 20, 147e157. http://dx.doi.org/10.1111/gcb.12341. dence from dietary and stable-isotope analyses. J. Exp. Mar. Bio. Ecol. 434e435, Sherrod, C., McMillan, C., 1985. The distributional history and ecology of mangrove 81e93. http://dx.doi.org/10.1016/j.jembe.2012.07.024. vegetation along the northern Gulf of Mexico coastal region. Contrib. Mar. Sci. Verges, A., Steinberg, P.D., Hay, M.E., Poore, A.G.B., Campbell, A.H., Ballesteros, E., 28, 129e140. Heck, K.L., Booth, D.J., Coleman, M.A., Feary, D.A., Figueira, W., Langlois, T., Sherrod, C., McMillan, C., 1981. Black mangrove, Avicennia germinans, in Texas: past Marzinelli, E.M., Mizerek, T., Mumby, P.J., Nakamura, Y., Roughan, M., van and present distribution. Contrib. Mar. Sci. 24, 115e131. Sebille, E., Gupta, A.Sen, Smale, D.A., Tomas, F., Wernberg, T., Wilson, S.K., 2014. Simas, T., Nunes, J.P., Ferreira, J.G., 2001. Effects of global climate change on coastal The tropicalization of temperate marine ecosystems: climate-mediated changes salt marshes. 139, 1e15. in herbivory and community phase shifts. Proc. R. Soc. B 281, 1e10. http:// Simpson, L.T., Feller, I.C., Chapman, S.K., 2013. Effects of competition and nutrient dx.doi.org/10.1098/rspb.2014.0846.

View publication stats